Instructions to use tencent/Penguin-VL-8B with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use tencent/Penguin-VL-8B with Transformers:

# Use a pipeline as a high-level helper
from transformers import pipeline

pipe = pipeline("text-generation", model="tencent/Penguin-VL-8B", trust_remote_code=True)
messages = [
    {"role": "user", "content": "Who are you?"},
]
pipe(messages)

# Load model directly
from transformers import AutoModelForCausalLM
model = AutoModelForCausalLM.from_pretrained("tencent/Penguin-VL-8B", trust_remote_code=True, dtype="auto")

Notebooks
Google Colab
Kaggle
Local Apps Settings

vLLM

How to use tencent/Penguin-VL-8B with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "tencent/Penguin-VL-8B"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "tencent/Penguin-VL-8B",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/tencent/Penguin-VL-8B

SGLang

How to use tencent/Penguin-VL-8B with SGLang:

Install from pip and serve model

# Install SGLang from pip:
pip install sglang
# Start the SGLang server:
python3 -m sglang.launch_server \
    --model-path "tencent/Penguin-VL-8B" \
    --host 0.0.0.0 \
    --port 30000
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:30000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "tencent/Penguin-VL-8B",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker images

docker run --gpus all \
    --shm-size 32g \
    -p 30000:30000 \
    -v ~/.cache/huggingface:/root/.cache/huggingface \
    --env "HF_TOKEN=<secret>" \
    --ipc=host \
    lmsysorg/sglang:latest \
    python3 -m sglang.launch_server \
        --model-path "tencent/Penguin-VL-8B" \
        --host 0.0.0.0 \
        --port 30000
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:30000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "tencent/Penguin-VL-8B",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Docker Model Runner
How to use tencent/Penguin-VL-8B with Docker Model Runner:
```
docker model run hf.co/tencent/Penguin-VL-8B
```

Penguin-VL-8B / configuration_penguinvl.py

lkeab

Transfer from pg-team/pg-vl-8b-hf

c8d1bab verified 3 months ago

raw

history blame contribute delete

1.73 kB

	"""PenguinVL model configuration."""

	import importlib.util
	import os.path as osp
	from typing import Optional, Dict, Any

	from transformers import PretrainedConfig, Qwen3Config

	try:
	from .configuration_penguinvl_encoder import PenguinVLVisionEncoderConfig
	except ModuleNotFoundError:
	spec = importlib.util.spec_from_file_location(
	"configuration_penguinvl_encoder",
	osp.join(osp.dirname(__file__), "configuration_penguinvl_encoder.py"),
	)
	configuration_penguinvl_encoder = importlib.util.module_from_spec(spec)
	spec.loader.exec_module(configuration_penguinvl_encoder)
	PenguinVLVisionEncoderConfig = getattr(
	configuration_penguinvl_encoder,
	"PenguinVLVisionEncoderConfig",
	)


	class PenguinVLQwen3Config(Qwen3Config):

	model_type = "penguinvl_qwen3"
	sub_configs = {"vision_encoder_config": PenguinVLVisionEncoderConfig}

	def __init__(
	self,
	vision_encoder: Optional[str] = None,
	vision_encoder_config: Dict[str, Any] = {},
	vision_projector_type: str = "mlp2x_gelu",
	use_token_compression: bool = True,
	image_token_index: int = -1,
	**kwargs,
	):
	super().__init__(**kwargs)
	self.model_type = "penguinvl_qwen3"

	self.vision_encoder = vision_encoder
	if vision_encoder_config is not None and not isinstance(vision_encoder_config, PretrainedConfig):
	vision_encoder_config = PenguinVLVisionEncoderConfig(**vision_encoder_config)
	self.vision_encoder_config = vision_encoder_config

	self.vision_projector_type = vision_projector_type
	self.use_token_compression = use_token_compression
	self.image_token_index = image_token_index